Measuring Services Innovation: Service Sector Innovation Index Hugo Hollanders UNU-MERIT (Maastricht University) Six Countries Programme (6CP) Workshop “Non-technical Innovations – Definition, Measurement & Policy Implications” 16-17 October 2008, Karlsruhe

Measuring services innovation: 2 approaches Industry-based Innovation in services is defined as innovation within service sector firms, i.e. firms active in NACE industries G to K Activity-based Services innovation covers any innovation activity with service like attributes that can occur in any part of the economy: manufacturing, agriculture, services or even informal parts of the economy

Activity-based approach: limited data availability Innovation surveys do not cover new business models and new customer interfaces Source: Forfás (2006), “Services Innovation in Ireland – Options for innovation policy”, a report commissioned by Forfás by CM International

New services: Community Innovation Survey Cross-sectional survey of all firms over 10 employees in EU Member States and EFTA countries over the period 2002-2004 (CIS-4 with over 60,000 respondents) / 2004-2006 (CIS-2006) Did your enterprise introduce: New or significantly improved goods? New or significantly improved services? CIS does not include all services sectors Aggregate CIS-4 results available from Eurostat, but not the differentiated results for goods/services (CIS-2006 results not yet available from Eurostat)

All sectors innovate by introducing new services In the Netherlands 55% of all product innovators introduced new services (29% in manufacturing, 69% in services)

New services: Innobarometer 2007 Quota survey of all firms over 20 employees in EU Member States and EFTA countries over the period 2005-2006 with over 5,000 respondents In the last two years, did your company perform any of the following innovative activities? Introduce new or significantly improved goods Introduce new or significantly improved services Innobarometer 2007 includes all services sectors

New service innovators more prominent in services sector In Europe 66% of all product innovators introduced new services, in particular in Knowledge-intensive services

Industry-based approach Industry-based approach is easier to use from a statistical point of view Availability of data (R&D, Community Innovation Survey) Comparability across countries (R&D) Synthesis consensus in literature: it is possible to use the same indicators to measure innovation in the manufacturing and services sectors, although some types of innovation are more common in services than in manufacturing (and vice versa) Service Sector Innovation Index (SSII) Methodology similar to Summary Innovation Index (European Innovation Scoreboard)

Service Sector Innovation Index: 23 indicators from CIS-4 grouped into 9 themes

Service Sector Innovation Index: sectors Manufacturing (NACE D) Services, including NACE G51 (wholesale) NACE I (transport, storage and communication) NACE J (financial intermediation) NACE K72 (computer and related activities) NACE 74.2 (architectural and engineering activities) NACE 74.3 (technical testing and analysis) KIBS (NACE 72+74.2+74.3) Services excl. KIBS

Composite innovation index For all sectors and indicators the data are For skewed variables: transformed using a square root transformation Normalised using the Min-Max method The SSII is calculated as the unweighted average of these normalised values

Methodological issues CIS-4 aggregate data are not available for all countries: a large number of countries could not be included Due to missing data some countries appear to under or over perform Some countries appear to under or over perform due to a national bias in how firms respond to the survey Some indicators are based on ‘yes’ or ‘no’ questions and do not account for differences in the intensity of an innovation activity Results need to be interpreted with care!!

Service Sector Innovation Index Luxembourg best performer in service sector innovation Different results for KIBS: Greece is best performer SSII correlates well with overall innovation performance Luxembourg best performer for Services. Greece best performer for KIBS, a result due to Greece scoring high in the effects from both product/process innovation and non-technological innovation. Results for Services and Services excl. KIBS are almost similar. SSII correlates well with overall innovation performance.

Service sector innovation is different Innovation in manufacturing relies on the accumulation of capabilities that require time to develop (both knowledge stocks and knowledge flows are vital) Innovation in services relies more on knowledge flows, permitting rapid ‘catching-up’ to best practice But some Member States with overall weak(er) innovation performance perform relatively well in service sector innovation (Estonia, Greece, Portugal)

Conclusions Can innovation in services be measured using indicators developed to measure (technical) innovation in manufacturing? Partly, as innovation in services is different as it relies more on knowledge flows permitting rapid ‘catching-up’ to best practice Can we compare service sector innovation across countries using CIS data? No, or only within groups of similar countries But, we can compare innovation within specific service sectors within each country

Recommendations - specific Eurostat/Member States should provide disaggregated results on the question if the enterprise introduced new or significantly improved goods or services Include questions on other types of services innovation, i.e. use of new business models and new customer interfaces Include all service sectors In particular NACE 73 (research and development) which is a key KIBS sector

Recommendations - general CIS data at sector level are missing for too many countries Increase sample size Increase response rates by introducing a standardized follow-up questionnaire Extend time dimension: is your firm using y? If yes, did you introduce y in: Last three years? Longer than three years ago?

Thank you For more details: Arundel, A., M. Kanerva, A. van Cruysen and H. Hollanders (2007), “Innovation Statistics for the European Service Sector”, Brussels: INNO Metrics 2007 Thematic Paper: www.proinno-europe.eu/metrics For questions or comments, please contact: Hugo Hollanders UNU-MERIT - Maastricht University PO Box 616, 6200 MD Maastricht, the Netherlands Tel +31 433884412 Fax +31 433884495 Website: www.merit.unimaas.nl Email: h.hollanders@merit.unimaas.nl